多尺度水平集SAR影像水体自动分割方法

被引:28
作者
徐川
华凤
眭海刚
陈光
机构
[1] 武汉大学测绘遥感信息工程国家重点实验室
关键词
多尺度分析; 合成孔径雷达; 水平集分割; OTSU; Gamma模型;
D O I
10.13203/j.whugis20120602
中图分类号
TN957.52 [数据、图像处理及录取];
学科分类号
摘要
针对高分辨率SAR影像的水体特性,提出了一种新的可精确、快速、自动完成高分辨率SAR影像水体分割的方法。该方法将多尺度分析技术与水平集理论相结合,在考虑全局信息的同时,也顾及到了局部信息。同时,为了满足SAR图像的分布特性,将Gamma统计模型引入到水平集能量函数中,利用OTSU算法来初始化零水平集。为了消除SAR影像中疑似水体的目标,对分割结果进行了后处理。实验表明,该方法在提高精度的同时,也大大提高了水体提取的效率。
引用
收藏
页码:27 / 31
页数:5
相关论文
共 8 条
[1]  
Influence of the noise model on level set active contour segmentation. P Martin,P Re?fre?gier,F Goudail,F Gue?rault. IEEE Transactions on Pattern Analysis and Machine Intelligence . 2004
[2]  
A level set method for oil slick segmentation in SAR images. Huang, B.,Li, H.,Huang, X. International Journal of Remote Sensing . 2005
[3]  
Separation between water and land in sar images using region-based level sets. Silveira, Margarida,Heleno, Sandra. IEEE Geoscience and Remote Sensing Letters . 2009
[4]  
A multiresolution stochastic level set method for Mumford-Shah image segmentation. Law, Yan Nei,Lee, Hwee Kuan,Yip, Andy M. IEEE Transactions on Image Processing . 2008
[5]  
Unsupervised extraction of flood-induced backscatter changes in SAR data using markov image modeling on irregular graphs. Martinis, Sandro,Twele, Andr,Voigt, Stefan. IEEE Transactions on Geoscience and Remote Sensing . 2011
[6]  
A run-based two-scan labeling algorithm. He, Lifeng,Chao, Yuyan,Suzuki, Kenji. IEEE Transactions on Image Processing . 2008
[7]  
Active contours without edges. Chan, Tony F.,Vese, Luminita A. IEEE Transactions on Image Processing . 2001
[8]  
A threshold selection method from gray-level histograms. Otsu N. IEEE Transactions on Systems Man and Cybernetics . 1979